From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.24) id 1AfhBy-0000UN-Re for user-mode-linux-devel@lists.sourceforge.net; Sun, 11 Jan 2004 07:07:54 -0800 Received: from smtp004.mail.ukl.yahoo.com ([217.12.11.35]) by sc8-sf-mx1.sourceforge.net with smtp (Exim 4.30) id 1AfhBy-0005rD-7o for user-mode-linux-devel@lists.sourceforge.net; Sun, 11 Jan 2004 07:07:54 -0800 From: BlaisorBlade Subject: Re: [uml-devel] Kernel mode fault at addr 0x48965000, ip0x400e51a4 References: <200401072025.02885.blaisorblade_spam@yahoo.it> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200401111334.04252.blaisorblade_spam@yahoo.it> Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: The user-mode Linux development list List-Post: List-Help: List-Subscribe: , List-Archive: Date: Sun, 11 Jan 2004 13:34:04 +0100 To: user-mode-linux-devel@lists.sourceforge.net Alle 21:33, sabato 10 gennaio 2004, Johannes Formann ha scritto: > At 20:25 Uhr +0100 07.01.2004, BlaisorBlade wrote: > >The fix maybe is not perfect but something must go in. Also, if that > > doesn't work, maybe the error is somewhere else (and triggered by this > > change, as easily happens). > > Has anyone found an workaround to get modules working with 23-1? Why do you say "modules"? I.e. do you get that problem in a special situation, like Jeff Chua that sees it only when he uses LVM? Does your UML uses an initrd to load some modules? > For the Simpel Patch doesn't work, and I think Jeff has change a few > things more in mem.c between 22-5 and 23-1. Jeff Chua posted 2 patches. The second was shorter BUT I'm *at all* sure it cannot do anything (by having read and understood it). I think you should apply his first change, or, as a workaround, disable klogd from inside UML or remove /dev/kmem. Your guest will work anyway (and inserting the proper fix will require more time). Also, on my systems, inside UML, klogd doesn't use /dev/kmem. --- linux/drivers/char/mem.c Wed Jan 7 02:58:37 2004 +++ uml/drivers/char/mem.c Wed Jan 7 03:07:52 2004 @@ -210,3 +210,2 @@ char * kbuf; /* k-addr because vread() takes vmlist_lock rwlock */ - p = (unsigned long) __va(p); if (p < (unsigned long) high_memory) { @@ -269,3 +269,3 @@ } - *ppos = __pa((void *) p); + *ppos = p; return virtr + read; Also, I've posted below (in this thread) a message which explains what is the bug (search for "[tracked down]" in the subject). So the debug work should be complete. The problem is that properly fixing it is maybe hard. -- cat <